Clustering for Collaborative Filtering Application in Web Recommendations

This paper mainly presents about the explosion of various contents generated on the web, Recommendation techniques have become necessary. Many types of data sources are used for recommendations and these data sources can be modeled in the form of graph. Aiming at providing the general limit on mining web graphs, we illustrate the generalization of different recommendation problems into our graph diffusion framework which can be utilized in many recommendation techniques on the World Wide Web including query suggestion and image recommendation.